University of Pennsylvania

ScholarlyCommons
Departmental Papers (Classical Studies) Classical Studies at Penn

2010

Literary Criticism
Joseph Farrell
University of Pennsylvania, jfarrell@sas.upenn.edu

Follow this and additional works at: http://repository.upenn.edu/classics_papers

Part of the Classics Commons

Recommended Citation (OVERRIDE)

“Literary Criticism.” In The Oxford Handbook of Roman Studies, ed. A. Barchiesi and W. Scheidel. Oxford: Oxford University Press,
2010: 176–187

This paper is posted at ScholarlyCommons. http://repository.upenn.edu/classics_papers/87

For more information, please contact repository@pobox.upenn.edu.
Literary Criticism
Abstract
Most of the literary theorists and critics of classical antiquity who are still studied today – Plato, Aristotle,
‘Longinus’, and a few others – are Greeks. The Romans, who by reputation came late to literature and lacked a
theoretical cast of mind, are not generally accorded a prominent place in the development of this discourse.
Indeed, few surviving Roman texts address as their main topic the business of literary criticism, at least as that
phrase is understood today. Nevertheless, the critical discussion of literature was a popular social activity
among the Roman elite and an obligation of the intelligentsia. Horace's Ars Poetica is the closest thing we have
to a Roman treatise on literary theory. The only actual treatises on poetics after Aristotle that might be relevant
to Roman literature have been found among the essays of the Greek philosopher and poet Philodemus of
Gadara.

Disciplines
Arts and Humanities | Classics

This book chapter is available at ScholarlyCommons: http://repository.upenn.edu/classics_papers/87

Literary Criticism

Oxford Handbooks Online

Literary Criticism
Joseph Farrell
The Oxford Handbook of Roman Studies
Edited by Alessandro Barchiesi and Walter Scheidel

Print Publication Date: Jun 2010

Subject: Classical Studies, Classical Poetry, Ancient Prose Literature
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199211524.013.0012

Abstract and Keywords

Most of the literary theorists and critics of classical antiquity who are still studied today –
Plato, Aristotle, ‘Longinus’, and a few others – are Greeks. The Romans, who by
reputation came late to literature and lacked a theoretical cast of mind, are not generally
accorded a prominent place in the development of this discourse. Indeed, few surviving
Roman texts address as their main topic the business of literary criticism, at least as that
phrase is understood today. Nevertheless, the critical discussion of literature was a
popular social activity among the Roman elite and an obligation of the intelligentsia.
Horace's Ars Poetica is the closest thing we have to a Roman treatise on literary theory.
The only actual treatises on poetics after Aristotle that might be relevant to Roman
literature have been found among the essays of the Greek philosopher and poet
Philodemus of Gadara.

Keywords: Plato, Aristotle, Longinus, literary criticism, Horace, Ars Poetica, literary theory, poetics, Roman
literature, Philodemus of Gadara

MOST of the literary theorists and critics of classical antiquity who are still studied today—
Plato, Aristotle, ‘Longinus’, and a few others—are Greeks. The Romans, who by
reputation came late to literature and lacked a theoretical cast of mind, are not generally
accorded a prominent place in the development of this discourse. Indeed, few surviving
Roman texts address as their main topic the business of literary criticism, at least as that
phrase is understood today. Nevertheless, the critical discussion of literature was a
popular social activity among the Roman elite and an obligation of the intelligentsia. For
professional writers, even from an early date, an explicitly critical perspective was de
rigueur, and almost every genre of Latin literature reflects this fact. But because so much
has been lost, we have to infer and extrapolate from fragmentary sources to understand
the full range of Roman activity. In addition, since this discourse overlaps with those of
grammar, rhetoric, and philosophy, all of which were more explicitly theorized and even

Page 1 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

better supported as cultural practices, we have to take into account their contributions as
well—without, however, allowing them to overwhelm our subject, as they constantly
threatened (and threaten) to do.

Most of what we know is told us by creative writers rather than by full-time theorists and
critics. Some of this information comes in the form of explicit statements of principle or
as criticism of other writers, and some of it has to be inferred from their own writerly
practice. What is explicitly stated does not always agree with what can be inferred. It is
fitting, then, that Horace's Ars Poetica is the closest thing we have to a Roman treatise on
literary theory. This work undoubtedly (p. 177) reflects the Roman reception of Greek
theory and criticism; but in places, ‘refracts’ might be a more accurate word. Not a
straightforward handbook, but a poetic representation of one, like Vergil's Georgics or
Ovid's Ars Amatoria, it is far from being the straightforward discussion that one might
like to have. In some ways, Horace's satires dealing with Lucilius (Serm. 1.4, 1.10) and
his literary epistles (Epist. 2.1, 2.2)offer more useful and representative perspectives. We
know of other verse treatises, earlier than Horace's, on different literary subjects. These
survive only in fragments, few but useful, and we shall meet them further on. Other
authors who make literary criticism an important element of their work include
Petronius, whose characters pronounce theoretical manifestoes and recite exemplary
compositions of tragedy, epic, Milesian tale, and other genres. Obviously, the irony that
permeates the Satyricon complicates our ability to draw definite inferences from such
material, not to mention rivalries within Nero's court. It is quite possible that Seneca's
tragedies are among Petronius' targets. But Seneca too shows himself an acute critic,
perhaps especially in his obiter dicta, such as the observation that Virgil wrote the
Georgics not so much to instruct farmers as to delight readers (Epist. 86.15)—a point that
has been lost on many another critic, ancient and modern.

The only actual treatises on poetics after Aristotle that might be relevant to Roman
literature have been found among the essays of the Greek philosopher and poet
Philodemus of Gadara. Titles include ‘On Poems’, ‘On Rhetoric’, ‘On the Good King
according to Homer’, and others. These essays fill some gaps in our knowledge of
Hellenistic theory while presenting a serviceable account of Philodemus' own views.
Philodemus (c.110–40/35 BCE), who was supported by the senator L. Calpurnius Piso
Caesoninus (cos. 58, cens. 50, father-in-law to Julius Caesar), lived on close terms with
Roman intellectuals, including some of the most important poets of the day (see e.g.
Armstrong 1995: 44). So the trove of essays found at Herculaneum—not all of them yet
published—should have a continuing influence on our understanding of the theoretical
underpinnings of Roman literature.

Page 2 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

More broadly, theory and criticism were implicated in the rather vaguely defined
discourse of ‘literary scholarship’. Most of the surviving material relevant to this
discourse was produced under the twin banners of Grammar and Rhetoric. Robert Kaster
(in this volume) points out that, for Roman scholars, ‘a desire to accumulate and
categorize plainly outstripped any desire to synthesize’—or, we may add, to theorize or
interpret. Much of the scholarship that has survived, such as the commentaries by
Donatus, Servius, et al., reflects the perspective of the grammaticus (Kaster 1989). These
works place a very high value on texts as collections of facts, or as vehicles for
communicating facts, about history, mythology, language, and so on. The approach has
little to do with the desire to construct a complete and coherent idea of how literature, or
a work of literature, is put together or of how to interpret it. Rather, the point is to impart
knowledge that might (or might not) shed light on almost any work of literature—or,
conversely, knowledge to which (p. 178) any work of literature might provide a
convenient point of access. This focus was well suited to elementary education, so that
the ars grammatica formed a basic part of the educated Roman's conception of literature.
Some aspects of grammar, such as the idea of canonical authors (on which more below),
did contribute to what we would recognize as literary theory and criticism, but others,
such as its remarkably doctrinaire character, are almost antithetical to a mature literary
sensibility.

The ars rhetorica was for more advanced students, and the theory of rhetoric was both
older and better developed than that of grammar. It was rhetoric that supplied much of
what passes, in the surviving evidence, for literary theory by providing a rationale for
producing as well as for judging actual literary performances (see Riggs by on ‘Rhetoric
as the Art of Discourse’ in this volume). According to this rationale, all texts are in
principle scripts for viva voce performance, and are judged accordingly. Today we must
remind ourselves just how important this aspect was for Roman readers: despite a
healthy book trade and a readership that extended throughout the empire, the
performance of literature remained a crucial social institution, and the ability to perform
oneself and to judge the performance of others was a defining mark of learning,
sensibility, and taste (Johnson 2000). For this reason, passages like the one in Quintilian
(IO 11.36–8) that explains how to deliver properly the opening lines of the Aeneid, are
valuable, not just as pieces of a ‘how-to’ manual, but as identifying the qualities that
make this very familiar passage an effective performance vehicle.

Still, rhetoric itself cannot account for all aspects of literary criticism that were of
interest to Roman writers and readers or that are of interest to us. The point needs
making because quite a few modern scholars have made the mistake of assuming that
ancient rhetoric does indeed tell the modern reader all that is needed about creating and
interpreting ancient literature. The importance of our grammatical and rhetorical sources

Page 3 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

is real, but they must not be allowed to obscure the ancient discourse of literary theory
and criticism per se, to which we now turn.

From the beginning Roman writers took a strong interest in evaluative criticism, canon
formation, philology, and literary history. The distinct outlines of a critical consciousness
appear in our earliest surviving literature. For Livius Andronicus, translating Homer's
Odyssey was an activity that required a philological approach to poetic language of the
sort that thrived in the Library of Alexandria (Mariotti 1952/1985). For Naevius, a
passage such as the enigmatic Gigantomachy ecphrasis (fr. 19 Morel—Buechner 1982)
seems indebted to allegorical exegesis in the style of the Pergamene critics, an important
critical tradition that has only recently come to be properly appreciated (Lamberton
1989; Struck 2004). Plautus identified and discussed his Greek sources in the prologues
to at least six of his comedies, and Terence's prologues are a rich and sophisticated
source of critical ideas and vocabulary. The tragic poet L. Accius (170–C.86 BCE)
composed a work, probably in prose, called Didascalica (conventionally ‘production
notices’, but the root meaning of ‘instructional materials’ may come into play) in at least
nine books on (inter alia) (p. 179) the Athenian and the Roman theater (frr. in Morel and
Buechner 1982). Another poem, of indeterminate length, written by one Volcacius
Sedigitus (c.100 BCE), was entitled ‘On Poets’ (De Poetis: Gellius, NA 15.24). Our sources
for the first 150 years of Latin literature are extremely spotty; but it is therefore
remarkable how much of the material that survives has to do with literary theory, history,
and criticism.

The impulses behind this activity are of course Hellenistic: the first Latin literature that is
visible to us bears an unmistakably Greek stamp. For us this is a fact. For the Romans it
may not have been so: they could easily have had, and probably did have, access to
earlier texts that appeared to them innocent of Greek influence. But the idea that the
Romans got all of their literature from Greece became received wisdom. It also became a
principle to which the literary critic might appeal. Ennius staked a claim to the
superiority of his Annales over the Bellum Poenicum of Naevius on the fact that Naevius
composed in uncouth Italic Saturnians, while Ennius had adapted the Homeric hexameter
to Latin (Ann. frr. 206–12 Skutsch). Ennius acted here as a highly interested critic, to be
sure. Nevertheless, his criticism of Naevius reflected both theoretical and practical
concerns, and it inaugurated a long tradition of critical and apologetic interventions. The
idea that Latin literature derives from Greek and is to be measured against Greek
became a basic element of Roman literary thought. For Horace, in fact, the idea actually
appears as an injunction: ‘look to your Greek models, night and day’ (AP 268–9).

In a certain sense, it is almost as if the overriding theoretical perspective regarded Latin

literature not only as deriving from Greek, but almost as a general effort to convert Greek

Page 4 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

literature, or the best of it, into Latin. A tendency, if not a rule, that informed this effort
was that any Greek work needed to be latinized only once. After Livius' Odusia no one
really needed to do the job again: in fact, a later version in hexameters survives (in a few
lines) only because it came to travel under Livius' name. Terence defends himself against
the charge, real or fictive, of adapting Greek plays that others had already made into
Latin versions. If two contemporaries simultaneously took on similar projects, it would be
for the reading public to decide between them. Cicero, writing to his brother about
Lucretius' poem, contrasts it favourably with the (otherwise unknown) Empedoclea of one
Sallustius, a text that he implies he could not finish reading (Q. Fr. 2.9.4). It may be that
part of Cicero's evaluation concerns which poem is the better rendition of Empedocles for
a Roman audience (Sedley 1998: 1–34). In his Brutus Cicero proceeds systematically
along these lines, comparing the defining characteristics of individual Roman orators to
those of specific Greek predecessors. He does not say that the Roman speakers had
deliberately modelled themselves on their Greek ‘counterparts’, as Lucretius and
Sallustius had done. Instead his comparisons, like Plutarch's syncrises, stand purely as
acts of interpretation. It could be that an unspoken system of this sort informed the way
in which the Augustan poets chose their projects, as if the various Greek authors
representing the different genres were arranged on a grid consisting of cells or slots
(Conte 1994: 116–17)—or, more concretely, of library shelves (Horsfall (p. 180) 1993)—
that remained available to all comers until successfully filled by a Roman equivalent.
Thus Virgil called his Eclogues ‘Syracusan’ or ‘Sicilian’ (= Theocritean) and his Georgics
‘Ascraean’ (= Hesiodic), Propertius called himself ‘the Roman Callimachus’, and Horace
boasted that he was the first to have shown Parian (= Archilochean) iambs to the Latin
people (Epist. 1.19.23–4) and that he followed this achievement in Aeolian (= Sapphic
and Alcaic) lyric.

By the later Augustan period this conceit seems to have run its course. The same Horace
who invokes it seriously was also capable of mocking it (Epist. 2.2.99–100). At about this
same time we encounter claims by this or that author that he deserves to stand alongside
earlier Roman as well as Greek writers. In this case there is less a question of latinizing
some Greek exemplar than of enriching the genre itself by whatever means. This attitude,
too, had earlier roots. In a passage quoted by Gellius (NA 15.24), the aforementioned
Volcacius Sedigitus lists Rome's ten best comic poets in descending order according to
his judgement, claiming that it is a controversial matter on which many have expressed
different opinions (fr. 1 Courtney). He does not include everyone who wrote comedies:
Livius Andonicus, for instance, is not named; Ennius appears in last place ‘out of respect
for his antiquity’. It seems obvious that Volcacius, in addition to making relative value
judgements, is also making the absolute judgement of which Roman comic poets belong
in the canon. He makes no comparison here between Greek and Roman poets, nor do we
know whether he promulgated canons for genres other than comedy, or whether his was

Page 5 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

the first Roman exercise in canon formation. But the critical impulse behind his work, and
its Hellenistic inspiration, are impressive. Volcacius followed the practice of Greek critics
by ranking specific poets as the most gifted exponents of a particular genre, the ones who
set the standard by which others would be judged. Poets like Horace (Carm. 3.30),
Propertius (2.34), and Ovid (Tr. 4.10) would later assert that their achievements had
earned them the right to be numbered with writers who excelled in their respective
genres. At the end of the Flavian period Quintilian gave this sort of criticism definitive
form, adding to the traditional Greek canon (IO 10.1.46–84) his own judgements
concerning the Romans (1.85–131). All that was left was for the idea of ‘classic’ authors
to crystallize, as it had done by the time of Marcus Aurelius. We will return to this idea
below.

In Volcacius' day the process of Roman canon formation was very much in flux; by the
time of Quintilian it was pretty well fixed. For Imperial poets this meant having to reckon
with towering Roman as well as Greek predecessors. It was no longer possible to refer
with disdain, as Ennius had done, to the insufficiently Hellenized efforts of the past, nor
was it necessary to defend an updated treatment of a familiar Greek work. Instead, any
new work had to vie with the most esteemed products of earlier Greek and Latin
literature.

Perhaps because Roman writers and critics constantly measured their own literature
against the standards first of earlier exemplars, they also had a reasonably well-
developed consciousness of literary history. The second-century poet and (p. 181) critic
Porcius Licinus (fr. 1 Courtney, apud Gellium NA 17.21.44) is the earliest writer to date
the arrival of literature at Rome:

With the Second Punic War, the Muse with winged gait
made war against Romulus' wild tribe.

Latin literature is thus not merely a product of Hellenism, but is to be numbered among the
manubial spoils taken during the conquest of a Mediterranean empire. Horace, who in one
famous passage (Epist. 2.1.156–62) echoed Porcius' conceit, boasted in another of being first ‘to
have brought home in triumph’ (deduxisse) to Italy the poetry of Sappho and Alcaeus (Carm.
3.30.13–14; cf. Epist. 1.19.26–33). The imagery of conquest is not accidental. It may allude to the
original conclusion (lost but plausibly conjectured: Skutsch 1985: 6, 553–63) of Ennius' Annales,
in which the poet may have likened his own conquest of the Greek Muses to that of his patron,
the Roman general Fulvius Nobilior, who conquered Ambracia and imported from there Rome's
first cult of the Muses. Porcius and Horace speak for the majority, though, in linking Rome's
acquisition of Greek literary culture to the prior conquest of Carthage, Rome's only important
rival as a world empire.
In any event, Roman critics traced the beginnings of their literature not to the mists of
time, but to datable events with definite military and political significance. As their

Page 6 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

chronological expertise improved, so did their grasp of literary history. Cicero, again in
the Brutus, developed explicitly the idea of periodization along with that of relativism in
literary evaluation. In doing so, he admitted that, over time, the best practitioners of a
given art will surpass the best of their forerunners. But he declared that it would be
wrong to condemn all earlier works as lacking the polish of the contemporary. Instead,
the connoisseur should understand the character of each generation and judge by that
standard. The idea is not original, but is borrowed from the history of art: Cicero refers to
Greek sculptors like Myron and Polyclitus, each with his own excellence and each the
best of his own generation (Brutus 70, 75).

This is a sophisticated view of the matter and a clear advance on the prevailing tendency
of Greek critics to divide literary history into ‘Homer’ and ‘later writers’. But it is not
clear how widely accepted the more advanced view was. Livy, in one of the most
frustrating passages of his history (27.37.1), informs us of Livius Andronicus' commission
to compose a hymn to Juno during a crisis in the Second Punic War, but he explicitly
refuses to quote the text because of its stylistic crudity. Cicero, at least, upheld his
relativistic theory in his practical criticism, declaring his admiration for Ennius against
the opinion of the literary avant garde (Tusc. 3.45), even though his own poetry has on
the whole more in common with the avant garde than with Ennius (Townend 1965).
Horace occupies a kind of middle ground: he dismisses his predecessor Lucilius as merely
a facile but careless versifier (Serm. 1.4.6–14), but admits that, had he lived in Horace's
own day, Lucilius would have observed a much higher standard (Serm. 1.10.64–74).

(p. 182) The notion of periodization appears in other guises as well. In Tacitus' Dialogus
de Oratoribus the dividing-line is the beginning of the Principate, and the great question
is whether oratory has maintained the social importance and the artistic standard that it
had achieved by the end of the Republic. Until then, important issues were decided by the
power of oratory in a climate of free speech; afterwards, with no scope for meaningful
action, oratory went into severe decline. Less pessimistically, Tacitus hails Trajan's
accession as the dawn of a new era in political and cultural life (Agr. 1–3). Tacitus' own
literary career and that of his friend Pliny do in fact begin at this time, and the following
century witnesses a distinctive literary culture, especially in prose but to a lesser extent
in poetry as well, before the dislocations of the third century have their effect. What is
perhaps most characteristic about second-century literature is precisely its literary-
critical and historical consciousness, which is characterized by a strong component of
archaizing and anti-classical revisionism (Holford-Strevens 2003: 193–225).

Roman criticism, then, regarded Roman literature as deriving from Greek, and judged
works of Roman literature largely on their success in adapting specific Greek models. It
borrowed the Hellenistic idea of the canon (Nicolai 1991) and fashioned a Roman canon

Page 7 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

on the model of the Greek one (Citroni 2005). But on what criteria did Roman critics base
their judgements that a specific effort was successful or not?

It is possible to distinguish between formalist and ethical criteria in Roman criticism,

even if ultimately the distinction is artificial, because the two strands are so thoroughly
intertwined. The grammatical and rhetorical category of ‘style’ is a basic formal
characteristic; and yet, it is notable that the stylistic choices of writers like Sallust and
Tacitus go closely with a distinct ethical persona, that of the highly judgemental,
disappointed idealist. By the same token, when Horace criticizes Lucilius' style he makes
the ethical or even moral observation that Lucilius simply shrank from the sheer effort
involved in writing well (Serm. 1.4.32–3). Nevertheless, formal and ethical categories
correspond approximately to two dominant concerns of modern theory and criticism,
which makes them, for us, useful heuristic categories.

From everything that has been said so far, it is clear that formal categories meant a lot to
Roman critics. Probably the most imposing of these categories was that of genre; and it is
also one of the most instructive. We cannot even name an ancient work that is likely to
have contained a well-developed theory of genres. Nevertheless, genres were used
constantly to categorize and evaluate literature—as we have already seen in the case of
canon formation. There as elsewhere, the simplest marker of any genre is metrical. Epic
poems were composed as a succession of dactylic hexameter lines (each of them called
epos in Greek), and epic poets composed such poems. Sub-genres were recognized as
well, and these might differ sharply from one another. Horace tells us that Varius was the
unrivalled master of epic in the ‘powerful’ (forte, acer) mode, while Virgil excelled in the
(p. 183) ‘gentle’ (molle atque facetum) mode (Serm. 1.10.43–5, written after the Eclogues
had appeared but before the Georgics had been made public or the Aeneid had begun to
take shape). To modern readers, the phrase ‘pastoral epic’ seems oxymoronic, because to
us epic means heroic, and pastoral and heroic poetry seem like entirely different genres.
We recognize the fact that both happen to be written in the ‘epic’ metre as technically
true, but relatively unimportant. To Roman critics, though, pastoral and heroic epic were
kindred sub-genres, distinguished by style and content, but related by a common metre.
Horace's hexameter poetry represents a third style, distinct from those of pastoral or
heroic epic and veering in the directions of conversation and epistolography. But Horace
begs the question of whether his hexameter compositions are really poetry at all (Serm.
1.4.39–65), muddying the waters still further.

At any rate, metre and subject-matter were in theory interrelated, constitutive elements
of genre, and genres remained true to themselves. Horace, again, states as much with
great clarity in the Ars Poetica (73–85), and the principle is echoed over and over by the
poets. This, however, is where prescriptive statements come into conflict with actual

Page 8 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

practice. It is not surprising that Roman poets transgress the boundaries of genre,
because no literary work of art can be permanently and exclusively assigned to any single
genre (e.g. Morson 1988). Roman literature is no different from any other in this respect.
What is perhaps different is that Roman writers, in their self-reflexive moments, embrace
the prescriptions of theorists and the strictures of critics, and suppose that their own
works ought to behave as perfect specimens of one single genre. But despite their guilty
consciences, they constantly flout these rules to produce works that are, from a generic
point of view, not merely hybrid but brilliantly and provocatively so.

Ovid is the champion of this schizophrenic pose. He constantly ‘worries’ (when he is not
actually boasting) about forcing his elegies to do what they were never meant to do. This
is amusing in itself, since elegy was to begin with a highly malleable genre without the
definite pedigree of genres like epic, tragedy, and comedy, so that it had always done
unprecedented things. But Ovid took this process to great lengths, and his expressions of
concern that he is transgressing the boundaries of genre seem intended to advertise the
fact (Farrell 2003). Feigned generic concern shows up in epic as well, where a succession
of poets fret over their decisions to introduce prominent female characters into this most
masculine of genres—in spite of the fact that women like Helen and Penelope had been
indispensible ‘elements’ of the genre since its Homeric beginnings (Hinds 2000). In both
these cases, poets acting as critics espouse a theory of genre that is widely at variance
with their usual practice, advancing this theory in a spirit of apparent self-reprehension,
and they do it with straight faces. These are passages of creative and critical
sophistication, and they force us to acknowledge the existence of an implied theoretical
discourse on genre more advanced than anything found in any surviving ancient treatise.

(p. 184) It is difficult to say what if any influence these sophistications had on those who
actually determined the Roman literary canon. It could be meaningful that Quintilian,
when evaluating Ovid as an elegist, ranks him (and Cornelius Gallus, who ‘founded’ the
genre) behind Tibullus and Propertius, finding him ‘too frivolous’ (just as he finds Gallus
‘too dour’, IO 10.1.93). Whether Quintilian is thinking only of the Amores, which are
comparable to the elegies of Tibullus and Propertius, or whether Ovid's frivolity
encompasses the generic experimentation embodied in the Ars Amatoria, the Heroides,
the Fasti, the exile poetry, and other works, we cannot say. It could also be meaningful
that the Ovidian work that Quintilian most admires is his generically unambiguous
tragedy, the Medea, and that he speaks only in passing of Ovid's masterpiece, the
unclassifiable Metamorphoses.

Quintilian's judgement on Ovid's elegies takes us from formal to ethical concerns: the
critic ranks the elegist as he does not because his poems were formally imperfect, but
because they were too silly. This is as if to say, they are definitely elegies (their metre

Page 9 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

tells us that much), but they are not sufficiently respectful of generic decorum. Just as in
Horace's judgement on Lucilius, ethical concerns are hard to separate from formal ones.
Many of the ideas involved could be represented in terms of dichotomies that pose
problems rather then stating simple truths. These include the ideas of the pleasing and
the useful (dulce, utile: Horace, AP 343) and the competing roles of genius and artistry
(ingenium, ars: Cicero, Q. Fr. 2.9.4; Ovid, Am. 1.15.13–14, 19; cf. Tr. 2.424) in producing
literature. Such concerns had of course been an important part of the earliest Greek
literary theory and criticism. They are rooted in the idea that a writer's work is a faithful
reflection of his character. For Plato and Aristotle this point is so fundamental that they
cannot imagine the same writer being successful in both tragedy and comedy. By
Hellenistic times versatile poets like Callimachus had disproved this theory many times
over, and Roman poets from Livius onward followed their lead (Farrell 2002). Here
especially theory failed to keep pace with practice: Quintilian remembers Callima-chus
simply as an elegist (IO 10.1.58), and the biographies of Roman poets, like those of the
Greeks, tend to treat a writer's life and his work as two sides of the same coin. A number
of poets, in different ways, insist that the work is not an accurate relection of the life.
Catullus chides Furius and Aurelius for supposing that his risqué themes bespeak a
debauched life (poem 16). Racy Martial (Epigr. 1.35) and prudish Pliny (Epist. 4.14.4)
both cite this poem to ward off criticism, as does Apuleius when defending himself in
court (Apol. 11). To a casual reader, these passages outline a simple commonplace; but in
the broader context of Roman literature and social life, they take on a more pointed
significance. And, in fact, with a lot at stake Ovid makes the ethical and moral gap
between his earlier work and the life that he has actually lived into a major theme of his
exile poetry.

In recent years literary criticism has increasingly abandoned the notion that the critic's
job is to distinguish better literature from worse and to explain the difference between
the two. In contrast, ancient criticism always involved value judgements (p. 185) and was
seldom if ever a matter of disinterested description. Nevertheless, the judgements that
went into producing the canonical literature of antiquity were deeply implicated in a
powerful social system, the workings of which have left a far clearer mark on the classical
literature that has survived, than have any abstract principles of literary excellence.

Traditionally, the very idea of ‘classical literature’ assumed that all works so labelled met
certain standards of excellence. But it is always worth remembering that this idea is in
origin a metaphor based on the Roman imperial tax code, and that the classic authors
are, by the terms of the metaphor, those who find themselves in a high bracket (M.
Cornelius Fronto, apud Gellium, NA 19.8.5; Curtius 1948, tr. 1953: 249–50; Citroni 2006:
204–11). The socio-political nature of this metaphor is the point. In antiquity no less than
today, a writer's reputation depended on many factors that had nothing to do with

Page 10 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

literary criticism as such. In the first place, Roman writers of all periods depended on
networks of social relations to find an audience (Starr 1987; White 1993). For most of the
Republic, poets tended to be men of lower social status who depended heavily on the
sponsorship of the socially powerful. In the late Republic, when men of more secure
position began devoting themselves to poetry, they generally continued attaching
themselves more or less closely to this or that patron. Writers of prose tended to be
members of the elite; therefore, instead of looking to patrons, they carefully adopted
distinctive positions with respect to other elite writers, for instance, in choosing the
addressees of individual works, in a way that mirrors the alliances that they were
constantly forming and then either nurturing or dissolving in their wider social and
political lives (Habinek 1998: 34–68). With the rise of the Principate, all writers found
themselves in a position of undeniable inferiority to one all-powerful patron, and state
sponsorship became the dominant factor in establishing a writer's reputation. It is true
that Rome was full of declaimers and reciters who might attract a following, and Martial,
for instance, could boastfully name specific bookshops that carried his own best-sellers
(Epigr. 1.2, 4.72, 5.16, 11.3). Moreover, Ovid, even though he complains in his exile
poems of being excluded from the Palatine Library (e.g. Tr. 3.1), nevertheless enjoyed
popularity in his lifetime and a spectacular Nachleben. But neither Martial nor any other
author earned money from the sale of his works, and Martial's success really derives from
being sponsored by the Flavian dynasty, just as his eclipse dates from the succession of
Nerva. As for Ovid, having been the most celebrated poet in Rome for most of his life was
a historical fact that his fall from grace could not undo. In reality, his career is an easily
explained exception to an iron rule, and the literate public, even if it played a role in
bringing a writer to the notice of the powerful, played a larger one in following the lead
and broadcasting the decisions of the patron class.

The question, then, is how literary patrons decided which writers to sponsor and which to
leave in eternal obscurity. Some, like Maecenas, were no doubt men of (p. 186) taste and
discernment, but we know little about the principles by which he and others discerned
good poets from bad. Indeed, Horace, who tells us most of what we know about this
process (Serm. 1.6), says practically nothing about aesthetic criteria. Instead, he speaks
of social and, again, ethical and even moral categories as the foundation of Maecenas'
little sodality. Horace does represent his powerful friend as relying on the advice of
others—in his own case, of Virgil and Varius—who would presumably be well qualified to
recommend a man on the basis of his talent as well as his character. But while Horace
does have something to say about his own literary ideals, he does not explicitly ascribe
them to Maecenas or to anyone else, preferring to characterize his relationship with his
patron and with these other men of letters as purely social and personal, rather than
essentially literary.

Page 11 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

Further Reading
There exist several surveys of ancient literary criticism. Inevitably and understandably,
these tend to concentrate on surviving prose treatises, most of which were written in
Greek (even if, like the treatise ‘On the Sublime’ wrongly ascribed to Longinus, they were
addressed to a Roman audience) and concern themselves mainly with grammatical and
rhetorical categories. These surveys remain helpful resources for students who wish to
situate the field of literary theory and criticism within the broader discourses of ‘literary
scholarship’. They include Grube 1965, Russell 1972, and Kennedy 1989. Russell 1996 is
also useful and compendious. Russell and Winterbottom 1972 contains a selection of
sources in English translation. Laird 2006 gives a different selection in new translations
and relates Greek and Roman criticism and theory to later developments. For the
evidence and interpretation of the theoretical foundations that are implied by the
imaginative and the critical work of Roman poets, see Schwindt et al. 2001.

References
ARMSTRONG, D. (1995). ‘The Impossibility of Metathesis: Philodemus and Lucretius on
Form and Content in Poetry’, in D. Obbink (ed.), Philodemus on Poetry: Poetic Theory in
Lucretius, Philodemus, and Horace. Oxford.

CITRONI, M. (2005). ‘The Concept of the Classical and the Canons of Model Authors in
Roman Literature’, in J. Porter (ed.), Classical Pasts: The Classical Traditions of Greece
and Rome. Princeton. 204–34.

CONTE, G. B. (1994). Genres and Readers: Lucretius, Love Elegy, Pliny's Encyclopedia, tr.
J. Solodow, rev. D. Fowler and G. W. Most. Baltimore.

COURTNEY, E. (1993), The Fragmentary Latin Poets. Oxford. Repr. with corrections,
Oxford, 2003.

CURTIUS, E. R. (1948; tr. 1953). European Literature in the Latin Middle Ages, tr. W.
Trask. New York.

FARRELL, J. (2002). ‘Greek Lives and Roman Careers’, in F. de Armas and P. Cheney (eds.),
Literary and Artistic Careers from Antiquity to the Renaissance. Toronto. 24–46.

——— (2003). ‘Classical Genre in Theory and Practice’, in R. Cohen and H. White (eds.),
Theorizing Genres II, New Literary History, 34/3: 383–408.

GRUBE, G. M. A. (1965). The Greek and Roman Critics. Toronto.

Page 12 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

HABINEK, T. (1998). The Politics of Latin Literature: Writing, Identity, and Empire in
Ancient Rome. Princeton.

HINDS, S. (2000). ‘Essential Epic: Genre and Gender from Macer to Statius’, in M. Depew
and D. Obbink (eds.), Matrices of Genre: Authors, Canons, and Society. Cambridge, Mass.
221–44.

HOLFORD-STREVENS, L. (2003). Aulus Gellius: An Antonine Scholar and his Achievement.

Rev. edn. Oxford.

HORSFALL, N. (2003). ‘Empty Shelves on the Palatine?’, Greece & Rome, 40: 58–67.

JOHNSON, W. (2000). ‘Toward a Sociology of Reading in Classical Antiquity’, AJP 121: 593–
627.

KASTER, R. (1989). Guardians of Language: The Grammarian and Society in Late Antiquity.
Berkeley.

KENNEDY, G. A. (ed.) (1989). ‘Classical Criticism’, in The Cambridge History of Literary

Criticism, vol. 1. Cambridge.

LAIRD, A. (ed.) (2006). Ancient Literary Criticism. Oxford.

LAMBERTON, R. (1989). Homer the Theologian: Neoplatonist Allegorical Reading and the
Growth of the Epic Tradition. Berkeley.

MARIOTTI, S. (1952). Livio Andronico e la traduzione artistica. Saggio critico ed edizione

dei frammenti dellʼ Odyssea. Urbino. Repr. 1985.

MOREL, W. and BUECUNER, C. (1982). Fragmenta Poetarum Latinorum. Leipzig.

MORSON, G. S. (1988). The Boundaries of Genre. Chicago.

NICOLAI, R. (1991). La storiografia nellʼeducazione antica. Pisa.

RUSSELL, D. A. (1972). Criticism in Antiquity. Oxford.

——— (1996). ‘Literary Criticism in Antiquity’, in S. Hornblower and A. Spawforth (eds.),

The Oxford Classical Dictionary.3 Oxford.

——— and WINTERBOTTOM, M. (1972). Ancient Literary Criticism. Oxford.

Page 13 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).
Literary Criticism

SCHWINDT, J. et al. (2003). LʼHistoire littéraire immanente dans la poesie latine. Huit
exposés suivis de discussions. Introduction by E. A. Schmidt. Fondations Hardt Entretiens
49. Geneva.

SEDLEY, D. (1998). Lucretius and the Transformation of Greek Wisdom. Cambridge.

SKUTSCH, O. (1985). The Annals of Quintus Ennius. Oxford.

STARR, R. (1987). ‘The Circulation of Literary Texts in the Roman World’, CQ 37: 213–23.

STRUCK, P. (2004). The Birth of the Symbol: Ancient Readers at the Limits of their Texts.
Princeton.

TOWNEND, G. (1965). ‘The Poems’, in T. A. Dorey (ed.), Cicero. New York.

WHITE, P. (1993). Promised Verse: Poets in the Society of Augustan Rome. Cambridge,
Mass.

Joseph Farrell
Joseph Farrell is Professor of Classical Studies at the University of Pennsylvania. His
main teaching and research interest is Roman poetry of the Republican and
Augustan periods.

Page 14 of 14

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy).